QAOD projects away question-aligned directions from answer representations to isolate domain-agnostic factuality signals, enabling efficient hallucination detection with top in-domain AUROC and up to 21% better OOD transfer.
Title resolution pending
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
representative citing papers
Fully Looped Transformer stabilizes looped training up to 12 iterations via distributed inter-loop signals and attention injection, improving downstream performance by up to 13.2%.
SMIXAE is a new mixture-of-autoencoders architecture that learns multidimensional manifolds directly from transformer activations, recovering known structures and identifying novel ones in Gemma 2 2B and 9B models.
Language model circuits show high within-task consistency and necessity but substantial overlap across tasks, making them less specific than assumed.
LLMs process negation using both attention-based suppression and constructive representation mechanisms (construction dominant), with late-layer attention shortcuts explaining poor accuracy on negation tasks.
Surprisal minimization over goal-directed alternatives generated by language models provides the strongest account of production choices in open-ended dialogue compared to uniform information density or length-based costs.
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
HORIZON creates a cross-domain, long-horizon user modeling benchmark from Amazon Reviews that tests generalization across time, domains, and unseen users, exposing gaps in sequential and LLM-based recommendation models.
MixRea benchmark reveals LLMs achieve at most 42.8% consistency on explicit-implicit reasoning tasks, with PRCP prompting proposed to recover overlooked relations.
A hybrid agentic architecture integrates knowledge-based physical verification tools into LLM-driven CAD design loops, producing more complex and functionally valid designs than prior agentic baselines.
ATD-Trans is a new geographically annotated Japanese-English travelogue dataset that reveals Japanese-enhanced models perform better on geo-entity translation while domestic Japanese locations remain harder to translate accurately.
ZAYA1-8B is a reasoning MoE model with 700M active parameters that matches larger models on math and coding benchmarks and reaches 91.9% on AIME'25 via Markovian RSA test-time compute.
FlexAttention supplies a compiler-driven interface that expresses common attention variants in a few lines of PyTorch and emits optimized kernels whose speed matches hand-written implementations.
Continual pre-training on a German medical corpus lets 7B models close much of the performance gap with 24B general models on medical benchmarks, though merging introduces some language mixing and verbosity.
LLMs compress concreteness into a consistent 1D direction in mid-to-late layers that separates literal from figurative noun uses and supports efficient classification plus steering.
LLM-based POS tagging outperforms traditional taggers on medieval Occitan, Catalan, and French, with fine-tuning and cross-lingual transfer providing the largest gains for under-resourced varieties.
citing papers explorer
No citing papers match the current filters.